PT  - JOURNAL ARTICLE
AU  - William Ricketts
AU  - Elan Tsarfati
TI  - The role of clinical history and investigations in the diagnosis and exclusion of tuberculosis
DP  - 2013 Sep 01
TA  - European Respiratory Journal
PG  - P499
VI  - 42
IP  - Suppl 57
4099  - http://erj.ersjournals.com/content/42/Suppl_57/P499.short
4100  - http://erj.ersjournals.com/content/42/Suppl_57/P499.full
SO  - Eur Respir J2013 Sep 01; 42
AB  - Introduction:Due to the limited sensitivity of definitive tests coupled with slow culture tuberculosis (TB) remains a clinical diagnosis, at least initially in those without sputum smear-positive disease. NICE guidance (2011) provides data on the role of clinical factors in diagnosis, but limited additional data on the role of investigations. We aimed to corroborate this data and investigate the additive effect of different tests and we investigate whether negative predictive values (NPV’s) can exclude TB. Our study was performed in a UK hospital serving a population with an annual TB incidence of ∼60/100,000.Methods:235 patients referred to TB clinic had history and investigations recorded in binary form i.e. present/absent or normal/abnormal. Of these 28 were diagnosed with active TB, the remainder with an alternative diagnosis.Sensitivity, specificity, positive predictive values (PPV’s), NPV’s and odds ratios were calculated using standard formulae for cough, fever, night sweats, weight loss, erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), white cell count (WCC), globulin, QuantiFERON and Mantoux.Results:Results are summarised in Table 1.Conclusions:Our data suggest that in the diagnosis of TB history is less important than investigations. However, as 15 (53.6%) of our diagnoses were clinical, with no laboratory confirmation our results may be a factor of clinician over reliance on investigations.